Gestational diabetes mellitus (GDM) and preeclampsia (PE) are associated with significant morbidity and mortality during pregnancy and risk for diabetes and cardiovascular disease after pregnancy. Biological mechanisms suggest alterations in specific cytokines (cytokines and growth factors) linked to inflammation, insulin resistance, and angiogenesis lead to GDM and PE. These cytokines may also serve as early markers or disease during pregnancy, and if present postpartum, critical markers for future disease. Currently, no single biochemical measure reliably identifies women at risk for these complex disorders during pregnancy, and those at risk for future disease. Likely, alterations in a specific combination of cytokines ('cytokine signatures') characterize GDM and PE. Limitations in technology and lack of prospective studies have prevented efficient and accurate quantification of cytokine combinations. Careful application of novel, multiplexed profiling techniques to quantify biologically important cytokines in women followed prospectively during pregnancy overcomes these limitations. We will test 3 hypotheses (in serum and urine): 1) Cytokine (TNF-alpha, IL-1beta, IL-6, MCP-1, IL-8) alterations identified early in pregnancy distinguish women who develop GDM or PE, and after pregnancy, characterize these same women; 2) Angiogenesis-related placental growth factors (PIGF, FGF-2) and an inhibitor (sFlt-1) are altered early in pregnancy among women who develop GDM and PE; 3) Disease-specific microarrays quantifying a set of pre-specified highly informative (aims 1, 2) cytokines can be developed, and this prospectively identifies incident GDM and PE. Our ongoing prospective pregnancy cohort permits excellent power and efficiency to test each aim. A multidisciplinary team with expertise in the epidemiology of diabetes and hypertension in pregnancy, placental biology, microarray technology, and bioinformatics will perform an innovative and cost-efficient study to understand the biology of GDM and PE, identify early markers for the conditions and suggest potential for interventions.